Rent Control is Problematic*

But there is a lot of specifics to discuss

A lot has been said about rent control, and the conventional economic informed viewpoint is mostly on the right track. There are many layers to explore here, so I want to get into it. The angles we must discuss are:

1. Supply and Demand
2. General Price Controls
3. Price Floors and Ceilings
4. Housing Policy
5. The Inventory Problem
6. Smart and Simple Housing Policy

First of all, this is an issue where people are going to bring up supply and demand. And in this case, the housing markets, is one of the few special cases where I think supply and demand is a good model. In most cases I don’t like the supply and demand model.

Why Supply and Demand is Almost Never The Best Model For Prices

Supply and Demand tends to lead to thinking about price in terms of “smooth curves”, or to be more mathematically specific: continuous, differentiable, and monotonic.

Mathematically reductive price curves:

These are the assumptions or simplifications people generally use when they want to apply the supply and demand model.

• Continuous: As price varies the supply or demand changes smoothly, without huge steps or gaps.
• Differentiable: We can measure how much one variable changes based on a change in the other variable.
• Monotonic: Higher prices always lead to more supply or less demand. Conversely, lower prices lead to less supply or more demand.

We will go through each of these simplifications in reverse order. The assumption of monotonicity is especially problematic given the way most modern goods are manufactured: they have fixed overhead costs and generally stable or declining marginal costs.

In a case with fixed overhead and stable or declining marginal costs, the goal is to sell as many units as you can to cover the overhead costs. Consider vehicle manufacturing. One reason making vehicles is so difficult, is because the overhead costs are so large. While the marginal costs of vehicle production can also be significant, they are generally much less of an issue. The big question in modern manufacturing and investing is: “how many units will I be able to sell”.

In most cases, if you can sell enough units, then you will recoup your investments and make a profit. So if a producer thinks they can sell a lot, it is also fairly likely they will be willing to sell at a lower price. This is in direct contrast to the conventional supply and demand model, which basically says that fewer are willing to produce a product or service at a lower price. Thus, the supply and demand model is mostly not very helpful.

Most of the time, when we think about prices, we are considering a environment where consumers already have an “anchor price” in mind, and so they use this as a benchmark to assess if something is a good deal. However, this “anchoring” of consumer expectations in a competitive environment, is a wholly different matter from how they would response to a shift in equilibrium market price, which is what the demand curve is supposed to quantify.

Again, this is just another example of how supply and demand are the least useful in the most interesting scenarios. The dynamics of pricing in extreme situations, will depend on what value a product has to consumers, if they may have to go without, and not what their general expectations are in a competitive and stable marketplace.

Finally, let’s talk about the critical situation where supply definitely does “slope” upward. This is when you hit systemic environmental resource limitations.

To discuss this more specifically, let’s consider a variation on the “traveling salesman problem”, but for an orchard with apples. For simplification, we will only consider a 2-dimensional variation of the problem: the picker must find the shortest path through the orchard to pick a given number of apples.

In this case, we can measure both the average cost of obtaining n apples, as well as the marginal cost of securing n+1 apples. If the apples are spread out very randomly in a 2 dimensional plane with distant outliers, then the this marginal cost will increase very smoothly. But if the apples are in a very regular pattern, then it will hardly cost more to pick all the apples compared to a portion of them.

But in both cases, once you have picked all the apples, the marginal cost of the next apple goes to infinity. And this is really the critical and interesting part of the problem. As utilization approaches systemic limits, costs increase much more rapidly than otherwise. This is something those familiar with computer science and algorithmic analysis should be very familiar with. Often computer systems can use resources very efficiently right up until the point that the resource is completely saturated, and then costs increase and efficiency breaks down. The ideal algorithm can use all of a resource with zero efficiency penalty. To be more specific, examples are filesystem management, cpu utilization, and much more.

A very generalized view of costs vs constraints is found in the domain of what was originally called “linear programming”, as well as more general mathematical optimization problems. In these problems an objective function is specified as well as constraints on a decision space. Techniques such as Lagrangian multipliers can be used to translate the system constraints into a modified objective function, but even with the straightfoward simplex algorithm, we learn how the constraints greatly affect optimality, and that they bind quickly. In fact, in a linear program, an optimal solution can always be found by checking the points where the most constraints possible are binding.

In the conventional view of supply and demand, we don’t get any of these interesting dynamics involving tradeoffs between resources and production variables, instead, costs are assumed to increase smoothly and continuously as you approach absolute resource limits.

Again, supply and demand is generally the least useful for the most interesting and important cases.

Why Supply And Demand Generally Applies to Housing Markets

Having said all that, supply and demand generally is a very good model for the housing market, and this is generally due to geographic constraints. For the most part, everyone must consume a minimum amount of housing(which can be artificially increased through excess regulation). Similarly, each unit of housing is geographically fixed to the particular market it is in.

For this reason, monopolies and monopsonies generally do not emerge in housing markets, although the ownership of housing may be concentrated along class divisions or even a cartel-like structure, it is a market that generally defies a high degree of consolidation and control.

Unlike manufactured goods, it is difficult or impossible for single producers to deliver massive quantities at lower prices. So generally the supply curve for housing is smooth and sloped upward.

What the Supply and Demand Model Actually Says About Price Controls

Many get this wrong, and claim that supply and demand states that price controls increase prices. As a simple matter of definitions, price ceilings like rent control cannot increase prices, except to infinity, meaning that the quantity is limited.

When applied to housing markets, what the supply and demand model would suggest is that price ceilings would lead to relocation. This could either be to areas with less demand or lower costs, or to areas without such price ceilings in place.

Note: Most Real World Rent Control Policies are Much More Nuanced Than Simple Price Caps

Most of the time, rent control does not actually set a ceiling on rental prices, it merely limits price increases. In these cases the costs are almost entirely from the landlord’s surplus, and do not increase prices. Once the construction is done, especially if developers were able to plan for an environment which limits price increases, then the costs have been accounted for at the existing price level. A limit on price increases merely cuts into the excess profits of landlords and speculators. So in these cases, rent control advocates have a completely valid point.

Despite this, even this form of rent control doesn’t actually increase the inventory available, it only makes costs more predictable for the tenant class of renters. So costs on new units may continue to rise aggressively, despite the policy otherwise achieve its goal of limiting landlord and land speculator surpluses by limiting price increases on the existing housing stock.

So it is really very anti-climactic. The one thing that this policy might have, is political viability. Housing policy is generally a political minefield, and if this style of consumer and renter protections can succeed politically, then it may be laudable, even if it underwhelming or the scope of impact limited to a set of long term tenants.

General Housing Policy Issues

This leads us to general housing policy issues, in which I will mostly just share my viewpoint. I think this is a point of tension between local and incumbent political interest, and more general interests.

Other options are land value taxes, public development of housing, and buffer stock policies(encouraging excess inventory to limit price shocks).

Historically, sometimes public housing projects were combined with progressive class targeted welfare. For housing this can be problematic because it doesn’t integrate well with the rest of the housing market, and concentrates groups of precarious tenants. This would be like taking all the struggling swimmers in a swimming lesson and putting them next to each other to try to rely on each for help. That is my explanation at least, there may be additional factors and other reasons behind this.

For this reason, if there is public development of housing, it should focus on the market’s needs, and not first as a welfare program. The goal should be to increase the inventory available at a reasonable price point, to make it easier for workers to have affordable quality options near jobs, and for businesses to have places where workers. While developers will want to speculate on land values, communities should naturally want to support a healthy and productive workforce, and that it is why it is so important that publicly developed housing should be sold at market prices, in the general market.

Housing Inventory Gets Contracted at The Bottom and Expanded at The Top

Even if the inventory of housing is good quality relative to its price, the tendency is for communities and politics to push out and end of life the affordable inventory, and expand and prioritize the expensive housing inventory.

This can lead to a housing affordability crisis when in fact, most housing is reasonably priced, or to exacerbate an across the board increase in housing prices. Often, I think both factors come into play.

There is a lot to discuss and consider on these issues, but if we do engage in public housing development, it should focus on the inventory problem, providing a greater amount of inventory at about the middle third of the distribution, that a new and developing workforce will be able to take advantage of.